How to applied 2d gaussian function

Hi,

I tried to modifed and applied this function: https://au.mathworks.com/matlabcentral/fileexchange/55033-fit-1d-and-2d-gaussian-to-noisy-data?s_tid=prof_contriblnk by @Manuel A. Diaz

I upload my code and data I don't know why the fit not working and the amplitude not in the center of the gaussain

I hope someone can help.

Thanks,

Answer 1

[Yongjian Feng]

Try to contact the auther of the fileexchange function first?

[Z998567]

Thanks, I'll try to contact him

Answer 2

[Matt J]

You could try this alternative gaussian fitting routine, and see if you get different results:

https://www.mathworks.com/matlabcentral/fileexchange/69116-gaussfitn?s_tid=srchtitle

[Z998567]

Hi Matt,

I try it, but I got an error with this "zdata: Mx1 vector of corresponding samples z(xdata)"

I convert my data to one row by "zdata = imageData(:);" but I got an error:

Error using gaussfitn (line 66)

Expected zdata to be an array with number of elements equal to 351.

Thanks

[Matt J]

Here's what I get,

load gauss

[X,Y]=ndgrid(1:length(imageData));

XY=[X(:),Y(:)];

params=gaussfitn(XY, imageData(:));

Local minimum possible.

lsqcurvefit stopped because the final change in the sum of squares relative to 
its initial value is less than the value of the function tolerance.

[D,A,mu,sig]=deal(params{:});

Zfit= D + A*exp( -0.5 *  sum( (sig\(XY.'-mu)).*(XY.'-mu) ));

imageFit=reshape(Zfit,size(imageData));

montage({imageData,imageFit})
caxis(4000*[0.7615    0.8126])

[Z998567]

Thanks Matt, it works but I need to plot the fir as an image like the attachement. Is there away with this function.

[Matt J]

Once you have the fitted parameters, you can plot it any way you wish.